Mobile device screens are increasingly being used as interfaces for touch events. Complex or multi-touch events, in particular, tend to generate a large amount of information that can overload the interface or the application. For example, a single touch event might describe the position, size, pressure and orientation of one or more fingers. Upwards of 100 touch events may be delivered to an application every second. One of the challenges that mobile device designers and engineers face is to minimize the latency between touch events and the availability of the touch event for use by an application.